Before coaching
Build the data engineering foundation coaching assumes.
This is the study path I give people who want to become data engineers but aren’t ready for coaching yet. It’s 32 hand-picked courses and books from ExpertEdge (Packt’s learning platform), ordered the way a working data engineer actually builds skills. Each learner does 16 or 17 core titles, depending on the platform they choose.
By Chris Gambill, Founder & Principal Architect, Gambill Data. Updated October 2026.
Disclosure: I earn a commission when you sign up with my code. I picked every title on this list myself and would recommend them either way.
Get access
Use code GAMBILL50 for $60 your first year.
The yearly plan is normally $120. The code does not apply to the monthly plan.
- Use code GAMBILL50 on the yearly plan for $60 your first year instead of $120. The code works on the yearly plan only, not monthly.
- Sign up at app.expertedge.org/sign-up. Titles link to their course pages once you’re signed in.
- Disclosure: I earn a commission when you sign up with my code. I picked every title on this list myself and would recommend them either way.
How to use this track
Do the core titles in order, and build as you go.
- Pace: the core track is about 100 to 110 hours of course time plus six small projects, roughly 14 to 16 weeks at 8 to 10 hours a week.
- Core vs optional: do every core title in order. Optional titles go deeper; skip them if time is tight.
- Phase 4 is a choice: pick the one platform you’re targeting (Databricks, Fabric, Snowflake, or AWS). Don’t try all four.
- Build as you go: every phase ends with a small project. A GitHub repo of those projects is worth more than any completion badge.
Phase 0 · 1 to 2 weeks
Engineer's toolkit
Data engineering is software engineering first, so the track starts with the tools every pipeline lives in.
| # | Title | Author | Hours | Core or optional | Why it’s here |
|---|---|---|---|---|---|
| 0.1 | Git Fundamentals: A Beginner's Guide to Managing Your Code | Bryan Krausen | 3.1 | Core | Branches, commits, and pull requests are table stakes on any data team. |
| 0.2 | The Software Developer's Guide to Linux | David Cohen, Christian Sturm | 5.0 | Core | Written for developers, not sysadmins. Covers the shell work data engineers actually do. |
| 0.3 | Python Fundamentals: Variables, Loops, Functions & More | Tyler Wall | 6.6 | Core | Current, focused Python basics without a data science detour. |
| 0.4 | Python Unit Testing Mastery: Complete Pytest and Mocking Guide | Faisal Memon | 3.3 | Optional | Testing is what separates pipeline code from notebook code. Strongly recommended. |
Phase project: A small Python script in a GitHub repo that reads a CSV, cleans it, and writes Parquet, with two pytest tests.
Phase 1 · 2 to 3 weeks
SQL and data modeling
SQL fluency and dimensional modeling are the skills most self-taught engineers skip, and the ones interviews and real projects lean on hardest.
| # | Title | Author | Hours | Core or optional | Why it’s here |
|---|---|---|---|---|---|
| 1.1 | SQL for Data Analytics (2025 edition) | Benjamin Johnston, Jun Shan, Haibin Li, and others | 5.6 | Core | Current SQL with window functions and advanced querying. Take the 2025 edition, not the 2019 one. |
| 1.2 | SQL Query Design Patterns and Best Practices | Steve Hughes, Dennis Neer, and others | 4.5 | Core | Readable, maintainable SQL: CTEs, patterns, and anti-patterns. |
| 1.3 | The Data Warehouse Toolkit | Ralph Kimball, Margy Ross | 10.1 | Core | The definitive text on dimensional modeling. The Gold layer of a medallion architecture still runs on these ideas. |
| 1.4 | Fundamentals of Analytics Engineering | Dumky De Wilde and others | 5.5 | Core | Bridges modeling and pipelines and introduces the analytics engineering mindset. |
Phase project: Design a star schema (one fact, three or four dimensions) for a public dataset and write the SQL to populate it, including a slowly changing dimension.
Phase 2 · 2 weeks
Pipelines, orchestration, transformation
This phase turns scripts into pipelines: ingestion with error handling, scheduling with dependencies, and SQL transformations under version control.
| # | Title | Author | Hours | Core or optional | Why it’s here |
|---|---|---|---|---|---|
| 2.1 | Data Ingestion with Python Cookbook | Gláucia Esppenchutz | 6.9 | Core | Practical ingestion patterns, including monitoring and catching errors, which most courses skip. |
| 2.2 | Apache Airflow Best Practices | Dylan Intorf, Dylan Storey, Kendrick van Doorn | 3.1 | Core | Orchestration concepts (DAGs, retries, idempotency) that transfer to Databricks Workflows and Fabric pipelines. |
| 2.3 | Data Engineering with dbt | Roberto Zagni | 9.6 | Core | SQL transformations with tests, documentation, and Git. The workflow most modern teams expect. |
Phase project: An Airflow DAG that ingests from a public API daily, lands raw files, and runs dbt models on top, with at least one dbt test.
Phase 3 · 2 to 3 weeks
Spark and the lakehouse
Distributed processing and open table formats are the core of modern data platforms, regardless of vendor.
| # | Title | Author | Hours | Core or optional | Why it’s here |
|---|---|---|---|---|---|
| 3.1 | Databricks Certified Associate Developer for Apache Spark Using Python | Saba Shah | 4.6 | Core | A solid, current PySpark and DataFrame API foundation that maps to a real certification. |
| 3.2 | Engineering Lakehouses with Open Table Formats | Dipankar Mazumdar, Vinoth Govindarajan | 6.9 | Core | Delta Lake, Iceberg, and Hudi side by side. Teaches why the lakehouse works, not just one vendor's buttons. |
| 3.3 | Data Engineering with Databricks Cookbook | Pulkit Chadha | 7.3 | Core | Hands-on Spark, Delta Lake, and pipeline recipes. |
Phase project: A Bronze, Silver, Gold pipeline in PySpark on Delta tables using Databricks Free Edition, with incremental loads and a MERGE for upserts.
Phase 4 · 2 to 3 weeks, choose one
Platform specialization
Pick the platform your target employers use. Snowflake learners should add Snowflake's own free training alongside these titles.
Databricks
| # | Title | Author | Hours | Core or optional | Why it’s here |
|---|---|---|---|---|---|
| 4.1 | Data Engineering with Azure Databricks | Tonya Chernyshova, Dmitry Anoshin, and others | 6.9 | Core | The newest Databricks title in the library. |
| 4.2 | Building Modern Data Applications Using Databricks Lakehouse | Will Girten | 4.1 | Core | Declarative pipelines, monitoring, and governance on Databricks. |
| 4.3 | Databricks Certified Data Engineer Associate: Practical Guide | Yogesh Raheja | 5.0 | Optional | Prep for the DE Associate exam if certification is a goal. |
| 4.4 | Mastering Azure Databricks for Data Engineers | 17.5 | Optional | Long, thorough PySpark and Spark SQL practice for people who want more reps. |
Fabric
| # | Title | Author | Hours | Core or optional | Why it’s here |
|---|---|---|---|---|---|
| 4.1 | The Definitive Guide to Microsoft Fabric | Christopher Maneu, Emilie Beau, and others | 12.6 | Core | The broadest current Fabric title: lakehouse, warehouse, pipelines, and governance. |
| 4.2 | Learn Microsoft Fabric | Arshad Ali | 5.6 | Optional | A second, more hands-on pass through the Fabric workloads. |
Snowflake
| # | Title | Author | Hours | Core or optional | Why it’s here |
|---|---|---|---|---|---|
| 4.1 | Data Modeling with Snowflake (2025 edition) | Serge Gershkovich | 7.2 | Core | Modeling for Snowflake specifically. Take the 2025 edition, not the 2023 one. |
| 4.2 | The Ultimate Guide to Snowpark | Shankar Narayanan SGS, Vivekanandan SS | 4.2 | Core | Python data engineering inside Snowflake. |
AWS
| # | Title | Author | Hours | Core or optional | Why it’s here |
|---|---|---|---|---|---|
| 4.1 | Data Engineering on AWS: The Complete Training | Ashish Prajapati | 18.9 | Core | End-to-end pipelines across the AWS analytics stack. Long, so it counts as the whole phase. |
| 4.2 | Ultimate AWS Data Engineering Bootcamp: 15 Real-World Labs | Siddharth Raghunath | 8.0 | Optional | Lab-heavy practice with Glue, Redshift, Airflow, and Spark. |
| 4.3 | Data Engineering with AWS Cookbook | Trâm Ngọc Phạm and others | 8.8 | Optional | Recipe-style reference for specific AWS tasks. |
Phase project: Rebuild the Phase 3 pipeline natively on your chosen platform, with a scheduled job and a simple dashboard on the Gold layer.
Phase 5 · 1 to 2 weeks
Quality, governance, streaming
These are the topics that separate someone who can build a pipeline from someone a team trusts to run one in production.
| # | Title | Author | Hours | Core or optional | Why it’s here |
|---|---|---|---|---|---|
| 5.1 | Driving Data Quality with Data Contracts | Andrew Jones | 3.4 | Core | Data contracts, ownership, and quality as an engineering practice, not an afterthought. |
| 5.2 | Apache Kafka Series: Learn Apache Kafka for Beginners v3 | Stéphane Maarek | 8.2 | Core | The standard Kafka introduction. Focus on the concepts, which carry over to Spark Structured Streaming. |
| 5.3 | Data Engineering Best Practices | Richard J. Schiller, David Larochelle | 9.2 | Optional | Architecture principles and SDLC practices. A good bridge toward senior-level thinking. |
| 5.4 | Hands-On CI/CD with GitHub Actions: Absolute Practical | Yogesh Raheja, Kulbhushan Mayer | 11.9 | Optional | General CI/CD with GitHub Actions. Not data-specific, but the mechanics transfer. |
| 5.5 | Data Observability for Data Engineering | Michele Pinto, Sammy El Khammal | 3.8 | Optional | Monitoring and alerting on pipelines and data freshness. |
| 5.6 | Practical Data Quality | Robert Hawker | 5.3 | Optional | Org-level data quality programs, useful context for senior roles. |
| 5.7 | Building a Modern Data Catalog with Microsoft Purview | Florian Detzel | 4.8 | Optional | Governance and cataloging for anyone heading into Azure or Fabric shops. |
Phase project: Add data quality checks and a written data contract to your Gold tables, and a small streaming source (Kafka or Structured Streaming) feeding Bronze.
What I’d skip for now
Hadoop, DP-203 prep, and dated Spark courses stay off the list.
Plenty of data engineering material still teaches patterns modern teams have moved past. I left these out on purpose.
- Hadoop-era big data: Hadoop, HDFS, and NiFi-centered courses.
- Superseded warehouse platforms: Azure SQL Data Warehouse and Synapse-first material, and prep for the retired DP-203 exam. Fabric and Databricks have taken over this ground.
- Older Spark courses: 2016 to 2022 Spark and PySpark courses. The concepts are fine, but the APIs and practices are dated.
- Data science detours: Python courses built around pandas visualization and machine learning. Useful, but not the shortest path to a data engineering role.
Where coaching picks up
Unity Catalog design, Spark UI reads, and senior-level interviews.
No course library covers the things that matter most for mid-level and senior roles. That’s what the Gambill Data Coaching Program is for.
- Unity Catalog and governance design
- Medallion architecture as a design discipline: naming, layering, and when to break the pattern
- Databricks Data Engineer Professional prep
- CI/CD for data: Databricks Asset Bundles, environments, and deployment pipelines
- Performance and cost: cluster sizing, partitioning and clustering choices, and reading a Spark UI
- Migration patterns: moving from SAS, SQL Server, or legacy warehouses to a lakehouse
- Career placement: resume, portfolio review, interview prep, and positioning for mid-level and above
Readiness checkpoint
- Finish the core titles and all six phase projects.
- Put your projects in a GitHub repo.
- Book a 20-minute Coaching Fit Call. We’ll review one project together and decide whether you’re ready to start the Gambill Data Coaching Program, or which gaps to close first.
You don’t have to finish everything before reaching out. If you’re stuck or want a second opinion on your plan, book a call anytime.